|
|||||||||
|
Thread Tools | Search this Thread |
January 31st, 2007, 07:22 AM | #1 |
Major Player
Join Date: Nov 2006
Location: Washington, USA
Posts: 213
|
I don't understand this part of HDV editing...
Hey guys... I am desperately seeking some help and guidance on this question I have about editing HDV that I just can't seem to find a detailed explanation of.
I just bought a Canon XH A1 HDV camcorder and I have Premiere Pro 2.0 and AE7. I also have a quad-core editing workstation with 4 GB RAM. I keep hearing people suggest that heavily editing native HDV (mpeg) in an application like PP2.0 will cause the video to degrade over the multiple generations of renders... is this correct? So what you are saying is, if I pull the untouched mpeg file natively into the timeline, perform some color correction, render a preview, then perform some transitions, titles, cuts, etc, then do another render preview... then save the project, my movie I created has lost resolution/quality over each step I mentioned? I thought PP2.0 made special "video preview" files that kept the original source footage untouched? And what happens if I go out and shoot a whole bunch of stock footage (quick clips) of various things and want to come home and razor up that footage in the timeline into smaller individual clips, then save each individual clip to their own mpeg file? Do I lose quality by doing this? And lastly... I have been reading on here all about Cineform AspectHD, and if possible, I'd rather edit natively since my computer can most definitely handle the heavy intensive processing, but if the major issue here is quality of picture and degradation of quality of picture, then there is no way I would sacrifice quality for $500 bucks. I appreciate any help or guidance on this matter, thanks so much for any light you can shed on this for me. |
January 31st, 2007, 10:34 AM | #2 |
Deke...
It really depends a lot on the kind of MPEG you're editting. The "long form" MPEG, the kind that's recorded by your HDV camera onto tape in the form of an m2t file, is highly compressed by the camera to make it fit onto the tape. The act of compression "throws away" a fair amount of image data, especially in the chroma. When you load this file into your NLE, your computer has to "reconstruct" the lost chroma data. Since the reconstruction is never a perfect duplicate of the original, some quality is lost. Software like Cineform, Sony YUV, or Raylight, make short form MPEG videostreams. The short form is a lot less lossy than long form, but, it also takes a lot more storage space. I beleive the conventional wisdom is that one can edit native m2t streams for cuts, only, If you're doing color correction or adding FX, the highest quality approach is to convert the m2t file to an intermediate like CFHD. |
|
January 31st, 2007, 12:17 PM | #3 |
Inner Circle
Join Date: Nov 2005
Location: Elk Grove CA
Posts: 6,838
|
My two cents:
I am a non-professional, advanced consumer. I seem to remember a thread or article reference here in DVi that actually showed what happened over successive rerenders of an HDV file. Just can't locate it now. You could definitely see a difference between render 1 and render 40. You can probably run the same test on a very small clip and see the difference. I don't feel I've lost a lot in editing HDV with transitions, color correction, and other effects, but I try to limit it to one generation. So if I am doing a project that has 5 or six distinct sections, I will do each one in its own sequence in PPro 2.0, then put it all together in a single final sequence for a render. In DV, I would actually render a new file. A year ago, every pro on this board was saying edit in intermediate. With the advent of Premiere 2.0, and Vegas 7.0 and dual core and more systems, that familiar refrain seems to have dissappeared somewhat. I'm guessing the reason is that there has been a discovery that the losses suffered aren't as significant as first thought, if care is taken. I also think there have been some issues with rights on the intermediate side that slowed that end up a bit. I am not sure where that is at this point.
__________________
Chris J. Barcellos |
January 31st, 2007, 05:17 PM | #4 |
Regular Crew
Join Date: Jan 2006
Location: Mexico City
Posts: 181
|
Deke,
You're right. None of the steps you described introduces any quality loss whatsoever, up until the moment you do the final render of your edited program. Intermediates will not improve your image quality one bit up until this stage, but they will help make your editing happen in real time on lesser machines. By the time you want to render you program file, the video will be recompressed into a new file, this is unavoidable, short of rendering your program file as an uncompressed file. If you intend to lay back onto hdv tape, even if you worked with intermediate files, at this point you won't be able to save your program from being recompressed into hdv and thus lose a little quality in the process. There are, however, a lot of options to render out your master for archival, and a better quality codec such as many of the intermediate offerings is a good option. Hope any of this helps. |
January 31st, 2007, 11:49 PM | #5 |
Trustee
Join Date: May 2004
Location: Knoxville, Tennessee
Posts: 1,669
|
Chris, there's something like that on the Cineform site - havent gone to look but its called "Quality tests" or similar
|
| ||||||
|
|